AMD scores big win with thousands of Instinct MI300x shipped to one of the world’s largest private cloud computing platforms
AMD’s Instinct MI300X is a beast of an AI accelerator – built on the company’s third-generation CDNA architecture and TSMC’s advanced 5nm and 6nm processes. It features 19,456 stream processors, 192 GB of HBM3 memory, 304 compute units, 1,216 matrix cores, and a thermal design power (TDP) of 750 watts. In one of the first benchmark tests, it absolutely destroyed Nvidia’s RTX 4090.
Vultra leading private cloud computing platform, has announced plans to integrate AMD’s powerful hardware with its scalable cloud infrastructure. The goal is to manage GPU-accelerated workloads more effectively and provide seamless performance for both data centers and edge computing.
“Innovation thrives in an open ecosystem,” said JJ Kardwell, CEO of Vultr. “The future of enterprise AI workloads lies in open environments that enable flexibility, scalability and security. AMD accelerators provide our customers with unparalleled cost-performance ratios. The balance between high memory and low power consumption advances sustainability efforts and gives our customers the ability to efficiently drive innovation and growth through AI.”
GPU-accelerated Kubernetes
Vultr’s partnership with AMD aims to provide enterprises with access to a world-class AI development environment. AMD’s architecture, combined with Vultr’s platform, will enable companies to work with open-source, pre-trained models, streamlining the development of AI projects.
“We are proud of our close partnership with Vultr, as the cloud platform is designed to manage high-performance AI training and inference tasks and deliver improved overall efficiency,” said Negin Oliver, corporate vice president of business development, Data Center GPU Business Unit , AMD.
“With the adoption of AMD Instinct MI300X accelerators and ROCm open software for these latest deployments, Vultr customers will benefit from a truly optimized system tasked with managing a wide range of AI-intensive workloads.”
This collaboration also optimizes Vultr’s infrastructure for GPU-accelerated Kubernetes clusters, capable of handling complex workloads globally. Vultr hasn’t revealed exactly how many Instinct MI300X units it has ordered, only that the initial cluster is in the “thousands.”
